Variable codings for: The Representational Effects of Communal Property

Conjoint and lab-in-the-field data was collected by the author. All data from those interventions has been made available in the replication materials. That data was then matched to the 2012 census of comunidades campesinas in Peru. Names of communities in the census were obtained through an information request of the Peruvian government. Identifying information of communities has been removed to protect anonymity of respondents. Data sources are included with each of the variables listed below. 

### comm_cens dataset ###

*comm_cens$ayni*

0: Community does not report preserving ayni
1: Community reports preserving ayni

Source: Questions P5801, P5802, and P5803 from CENAGRO 2012.

---

*comm_cens$year_rec* 

Min: 1921, Max: 2009

Source: Directory of Comunidades Campesinas (2009)

---

*comm_cens$none*

0: Integrated into market
1: Not integrated into market

Source: Questions P02601, P02602, P02603, P02604, P02605, P02606, P02607, P02608, P02609 from CENAGRO 2012. If any are equal to 1, code variable as 0. If *all* are equal to 0, code variable as 1. 

---

*comm_cens$market*

0: Not integrated into market
1: Integrated into market

Source: See comm_cens$none.

---

*comm_cens$int*

0: Market integration and no ayni
1: Ayni and market integration
2: No market integration and ayni
    
Source: comm_cens$none + comm_cens$ayni

---

*comm_cens$quechua*

0: Community primarily speaks another indigenous language
1: Community primarily speaks Quechua

Source: Question P0140201 from CENAGRO 2012.

---

*comm_cens$aymara*

0: Community primarily speaks another indigenous language
1: Community primarily speaks Aymara

Source: Question P0140201 from CENAGRO 2012.

---------------------------------------------

## Lab-in-the-field experiment ("data") ##

*data$campesino*

0: Partner is from Cusco
1: Partner is from community

---

*data$out1* Amount given by player 1 (in Peruvian soles)

0, 1, 2, 3

---

*data$out2* Dummy measure for giving something vs. nothing

0: Gave nothing
1: Gave something

---

*data$compliers1* 

0: Answered no or don't know to manipulation check
1: Answered yes to manipulation check

---

*data$compliers2* 

0: Answered don't know to manipulation check
1: Answered yes to manipulation check

---

*data$ayni*

0: Community reports preserving ayni
1: Community does not report preserving ayni

Source: Questions P5801, P5802, and P5803 from CENAGRO 2012.

---

*data$comunidad_full*

Community name (character string)

---

*data$distrito*

District/municipality name (character string)

---

*data$title_date*

Date of community title

0: Pre-1960
1: Post-1960
2: No title

---------------------------------------------

## Candidate dataset ("cand")##

Source: https://infogob.jne.gob.pe/

---

*cand$com_cand*

Number of community-member candidates

---

*cand$more_than_one*

0: Only one community candidate
1: More than one community candidate

---

*cand$all_ayni*

0: Not all communities preserve ayni
1: All communities preserve ayni

---------------------------------------------

## Conjoint dataset ("conj")##

*conj$ratings*
Evaluation of profiles (rating)

1: Strongly oppose
5: Strongly support

---

*conj$outcome*
Evaluation of profiles (discrete choice)

0: Did not choose candidate
1: Chose candidate

---

*conj$tu_comunidad*

1: Candidate is from your community

---

*conj$una_comunidad*

1: Candidate is from a community

---

*conj$water*

1: Candidate's primary platform issue is water

---

*conj$electricity*

1: Candidate's primary platform issue is electricity

---

*conj$ed*

1: Candidate's primary platform issue is education

---

*conj$male*

1: Candidate is a male

---

*conj$pol_party*

1: Candidate belongs to political party

---

*conj$reg_mov*

1: Candidate belongs to regional movement

---

*conj$id*

Respondent ID

---------------------------------------------

## Merged data: conjoint and lab-in-the field (data2)##

*data2$lab_in_field*

0: T-stat of municipal-level lab-in-field effect < 1.6
1: T-stat of municipal-level lab-in-field effect > 1.6


